Optical reception device

Abstract

[Problem] It is possible to cancel out beat noises in a WDM signal to obtain a high-quality signal component, and a configuration thereof is achieved with low mounting costs. [Solution] An optical reception device converts a WDM signal rs(t) of an optical signal to electrical signals d(t)1 to d(t)n expressed as complex numbers of orthogonal phases, cancels out beat noises from the electrical signals, and then demodulates the signals to obtain signals D1 to Dn of a transmission source. This optical reception device includes an absolute value calculation units 31a to 31n that generate a signal component including a beat noise component through absolute value squaring calculation processing for squaring an absolute value of the electrical signal, scaling units 32a to 32n that multiply the generated beat noise component by a proportional constant to restore beat noise associated with n types of wavelengths in the electrical signal, and subtraction units 33a to 33n that subtract the n type of beat noises restored from the electrical signal to cancel out then types of beat noises included in the electrical signal.

Claims

1. An optical reception device configured to select optical signals having same wavelength as local light with a predetermined wavelength, phases of the optical signals are orthogonal, to each wavelength of wavelength division multiplexing (WDM) signal in which optical signals having a plurality of different wavelengths are multiplexed, convert the selected optical signals with their orthogonal phases into electrical signals and expressed as complex numbers, cancel out beat noises from the electrical signals, and demodulate the electrical signals to obtain signals of a transmission source, the optical reception device comprising: a signal processing circuit configured to: obtain an absolute value of complex number expressed by each electrical signal and perform absolute value squaring calculation processing for squaring the absolute value, to generate a signal component including a beat noise component; multiply the beat noise component by a proportional constant to restore beat noises associated with a plurality of wavelengths in each electrical signal; and subtract the restored beat noises from each electrical signal to cancel out the beat noises included in each electrical signal.

2. The optical reception device according to claim 1, wherein the signal processing circuit is further configured to: feedback a signal after the beat noise is canceled out, wherein the feedback is repeated a plurality of times.

3. The optical reception device according to claim 1, wherein the signal processing circuit is further configured to: demodulate a signal after the beat noise cancellation is processed; compare a voltage level of the demodulated signal to a predetermined threshold value to determine a transmission source signal component with no beat noise of the transmission source; generate an appropriate transmission source signal with no beat noise of the transmission source according to the transmission source signal component; perform the absolute value squaring calculation processing on the generated transmission source signal to generate an accurate beat noise component, and generate a signal component including the generated accurate beat noise component; synchronize the signal component including the accurate beat noise component with each electrical signal, and multiply the accurate beat noise component by a proportional constant to restore a beat noise associated with the generated transmission source signal; and cancel out the beat noise included in each electrical signal on the basis of the restored beat noise.

4. The optical reception device according to claim 1, wherein the signal processing circuit is mounted in a semiconductor chip.

5. The optical reception device according to claim 1, wherein the transmission source signal is transmitted in discrete voltage patterns.

Description

BRIEF DESCRIPTION OF DRAWINGS

(1) FIG. 1 is a block diagram illustrating a configuration of an optical reception device according to an embodiment of the present invention.

(2) FIG. 2 is a block diagram illustrating a configuration of a canceler unit in the optical reception device according to the embodiment.

(3) FIG. 3 is a block diagram illustrating a configuration of a canceler unit according to Modification Example 1 of the embodiment.

(4) FIG. 4 is a block diagram illustrating a configuration of a canceler unit according to Modification Example 2 of the embodiment.

(5) FIG. 5 is a block diagram illustrating a configuration of an optical reception device using a colorless reception scheme using balanced detection of the related art.

(6) FIG. 6 is a block diagram illustrating a configuration of an optical reception device using a digital coherent WDM transmission scheme using single-type detection of the related art.

DESCRIPTION OF EMBODIMENTS

(7) Hereinafter, embodiments of the present invention will be described with reference to the drawings. However, corresponding components in all of the drawings of the present specification are denoted by the same reference signs, and descriptions thereof will be omitted appropriately.

Configuration of Embodiment

(8) FIG. 1 is a block diagram illustrating a configuration of an optical reception device 10C according to an embodiment of the present invention.

(9) The optical reception device 10C illustrated in FIG. 1 is of a single ended type that is the same as that illustrated in FIG. 6, and includes n optical hybrid units 11a to 11n, n local light sources 12a to 12n, PDs 13I connected to I ports on the output side of the respective optical hybrid units 11a to 11n, and PDs 13Q connected to Q ports on the output side of the respective optical hybrid units 11a to 11n.

(10) Further, the optical reception device 10C includes a multi-channel compatible interference canceler unit (also referred to as a canceler unit) 30, which is a characteristic element of the embodiment, and demodulation units 15a to 15n that are the same as those illustrated in FIG. 5.

(11) The canceler unit 30 includes n absolute value calculation units 31a, 31b, . . . , 31m, and 31n, n scaling units 32a, 32b, . . . , 32m, and 32n, and n subtraction units 33a, 33b, . . . , 33m, and 33n, as illustrated in FIG. 2. The canceler unit 30 is configured by forming a signal processing circuit constituting the canceler unit 30 on one substrate, or is configured by mounting a signal processing circuit constituting the canceler unit 30 in one semiconductor chip.

(12) The optical hybrid units 11a to 11n illustrated in FIG. 1 extract an optical signal having the same wavelength as that of the local light by causing an input WDM signal r.sub.s(t) and the local light having the first to n-th wavelengths of the respective local light sources 12a to 12n to optically interfere in an optical waveguide, and output the extracted signal light from the I port and the Q port, with phases of the optical signal being made orthogonal. Optical signals of an I channel from the I ports of the optical hybrid units 11a to 11n are converted to electrical signals in the PD 13I, and optical signals of the Q channel from the Q ports are converted to electrical signals in the PD 13Q.

(13) The above WDM signal r.sub.s(t) is expressed by Equation (1) below.

(14) $\begin{array}{cc}\left[\mathrm{Math}.1\right]& \\ r_S\left(t\right)=m_S\left(t\right)e^{j{\omega }_{\mathrm{cs}}}+\underset{i=1}{\overset{N-1}{.Math.}}{m}_i\left(t\right)e^{j{\omega }_{\mathrm{ci}}}& \left(1\right)\end{array}$

(15) Here, a first term on the right side indicates signal components selected by local light from the n local light sources 12a to 12n, and indicates n signal components having a frequency matching a frequency of each local light. A second term on the right side indicates a signal including a beat noise component.

(16) The electrical signals of the I and Q channels for each wavelength obtained by converting the optical signals by the respective optical hybrid units 11a to 11n are input to the canceler unit 30 illustrated in FIG. 2 in a state in which the electrical signals are plotted on a complex plane with orthogonal I and Q axes, that is, as electrical signals that are complex values.

(17) For the electrical signal input to the canceler unit 30, a signal associated with the first wavelength output from the optical hybrid unit 11a is set as a first electrical signal d(t)1, a signal associated with the second wavelength output from the optical hybrid unit 11b is set as a second electrical signal d(t)2; . . . , a signal associated with the m-th wavelength output from the optical hybrid unit 11m is set as the m-th electrical signal d(t)m, and a signal associated with the n-th wavelength output from the optical hybrid unit 11n is set as the n-th electrical signal d(t)n.

(18) In the canceler unit 30, the absolute value calculation unit 31a obtains an absolute value of the complex number, which is the first electrical signal d(t)1, and performs absolute value squaring calculation processing for squaring the absolute value to generate a signal component including a beat noise component as a baseband signal. Similarly, the other absolute value calculation units 31b to 31n perform the absolute value squaring calculation processing on each of the second to n-th electrical signals d(t)2 to d(t)n to generate a signal component including a beat noise component as a baseband signal.

(19) For example, the baseband signal generated through the absolute value squaring calculation processing is expressed as in Equations (2a) and (2b) below. Here, only an I channel component is expressed. A Q channel is omitted because the Q channel is the same.

(20) $\begin{array}{cc}\left[\mathrm{Math}.2\right]& \\ d\left(t\right)={.Math.{\mathrm{Ae}}^{j{\omega }_{\mathrm{cs}}}+m_S\left(t\right)e^{j{\omega }_{cs}}+\underset{i=1}{\overset{N-1}{.Math.}}{m}_i\left(t\right)e^{j{\omega }_{\mathrm{ci}}}.Math.}^2& \left(2a\right)\\ \approx 2A{m}_S\left(t\right)+\underset{i=1}{\overset{N-1}{.Math.}}{.Math.m_i\left(t\right).Math.}^2& \left(2b\right)\end{array}$

(21) In the above Equation (2a), a first term on the right side is local light, a second term on the right side is the same signal component as in the first term of Equation (1) above, and a third term on the right side is a signal including the same beat noise component as in the second term of Equation (1).

(22) Equation (2b) is expressed as being approximately equal because some of various signal terms calculated in the absolute value squaring calculation processing of Equation (2a) are negligible. A first term on the right side of the approximately equal sign indicates a signal component, and a second term on the right side indicates a beat noise component that is to be canceled.

(23) The scaling units 32a to 32n multiply each beat noise component after the absolute value squaring calculation processing by a proportional constant to restore the beat noise of each of the electrical signals d(t)1 to d(t)n. These restored beat noises are input to the respective subtraction units 33a to 33n.

(24) For the subtraction units 33a to 33n, for example, the subtraction unit 33a subtracts the n types of restored beat noises from the first electrical signal d(t)1, and cancels out all (n types of) beat noises included in the first electrical signal d(t)1. That is, all the beat noises output from the respective scaling units 32a to 32n are subtracted from all the beat noises of the WDM signal r.sub.s(t) included in the first electrical signal d(t)1 such that all the beat noises are canceled out.

(25) Through this cancellation, the first signal D(t)1 that is a signal component of the first electrical signal d(t)1 is output from the subtraction unit 33a, a second signal D(t)2 that is a signal component of the second electrical signal d(t)2 is output from the subtraction unit 33b, . . . , an m-th signal D(t)m that is a signal component of the m-th electrical signal d(t)m is output from the subtraction unit 33m, and an n-th signal D(t)n that is a signal component of the n-th electrical signal d(t)n is output from the subtraction unit 33n.

(26) The respective electrical signals d(t)1 to d(t)n after the beat noises are canceled out are expressed by Equations (3a) and (3b) below.

(27) $\begin{array}{cc}\left[\mathrm{Math}.3\right]& \\ D\left(t\right)=d\left(t\right)-\gamma \underset{i=1}{\overset{N-1}{.Math.}}{.Math.d\left(t\right).Math.}^2& \left(3a\right)\\ \approx 2A{m}_S\left(t\right)+\underset{i=1}{\overset{N-1}{.Math.}}{.Math.m_i\left(t\right).Math.}^2-\gamma .Math.4A^2\underset{\left(i\ne S\right)}{\underset{i=1}{\overset{N-1}{.Math.}}}{.Math.m_i\left(t\right).Math.}^2& \left(3b\right)\end{array}$

(28) D(t) in Equation (3a) above indicates a first signal D(t)1 obtained by subtracting, from the first electrical signal d(t)1, all the beat noise components restored from the electrical signal d(t)1. Similarly, D(t) indicates respective signals D(t)2 to D(t)n obtained by subtracting, from the respective other electrical signals d(t)2 to d(t)n, all of the beat noise components restored from the respective electrical signals d(t)2 to d(t)n.

(29) A right side of Equation (3b) above indicates that the beat noise component of a second term is canceled out with the beat noise component of a third term, and only the signal component of the first term is left. The proportional constant γ in the third term is appropriately selected in order to perform this cancellation, in other words, to cancel out the beat noise component. That is, γ is a parameter that is adjustable for cancellation, and is provided to be adjustable in the subtraction units 33a to 33n.

(30) The respective signals D(t)1 to D(t)n obtained in this manner are demodulated by the demodulation units 15a to 15n illustrated in FIG. 1, and first to n-th signals D1 to Dn of the transmission source are obtained.

Operation of Embodiment

(31) Next, an operation of the optical reception device 10C according to the embodiment will be described. A representative case in which the first signal D(t)1, which is a signal component, is obtained from the WDM signals r.sub.s(t) will be described herein.

(32) When the WDM signal r.sub.s(t) illustrated in FIG. 1 is input to the optical hybrid unit 11a, an optical signal having the same wavelength as the local light having the first wavelength from the local light source 12a is extracted from the WDM signals r.sub.s(t). Further, in the optical hybrid unit 11a, first optical signals of the I and Q channels in which the extracted signals are orthogonal are output from the I and Q ports. The first optical signal of the I channel output from the I port is converted to an electrical signal in the PD 13I, and the optical signal of the Q channel output from the Q port is converted to an electrical signal in the PD 13Q.

(33) The electrical signals of the I and Q channels for each wavelength after the conversion are input to the canceler unit 30 illustrated in FIG. 2 as the first electrical signal d(t)1, which is a complex value in an orthogonal state.

(34) The input first electrical signal d(t)1 is subjected to an absolute value squaring calculation processing by the absolute value calculation unit 31a, and a signal component including the beat noise component is generated. Then, in the scaling unit 32a, the beat noise component is multiplied by the proportional constant, thereby restoring all of the beat noises included in the electrical signal d(t)1. These restored beat noises are input to the respective subtraction units 33a to 33n.

(35) In the subtraction unit 33a, all of the above restored beat noises are subtracted from the first electrical signal d(t)1, thereby canceling out all the beat noises included in the first electrical signal d(t)1. Through this cancellation, the first signal D(t)1, which is a signal component of the first electrical signal d(t)1, is output. This first signal D(t)1 is demodulated by the demodulation unit 15a illustrated in FIG. 1, so that the first signal D1 of the transmission source is obtained.

Effects of Embodiment

(36) An effect of the optical reception device 10C according to the embodiment will be described. This optical reception device 10C extracts optical signals having the same wavelength as the local light having a predetermined wavelength from the WDM signal r.sub.s(t) in which optical signals having n types of different wavelengths are multiplexed, with phases of the optical signals being made orthogonal, converts the extracted optical signals having the orthogonal phase to electrical signals d(t)1 to d(t)n expressed as a complex number, cancels out beat noise from the electrical signals d(t)1 to d(t)n, and then demodulates the signals to obtain the signals D1 to Dn of a transmission source.

(37) A characteristic of the embodiment is that the optical reception device 10C includes the canceler unit 30 including the absolute value calculation units 31a to 31n, the scaling units 32a to 32n, and the subtraction units 33a to 33n.

(38) The absolute value calculation units 31a to 31n obtain the absolute values of the complex numbers, which are the electrical signals d(t)1 to d(t)n and perform the absolute value squaring calculation processing for squaring the absolute values to generate the signal components including the beat noise components.

(39) The scaling units 32a to 32n multiply the generated beat noise components by the proportional constant to restore the beat noises associated with the n types of wavelengths in the electrical signals d(t)1 to d(t)n.

(40) The subtraction units 33a to 33n subtract the n types of restored beat noises from the respective electrical signals d(t)1 to d(t)n, and cancel out the n types of beat noises included in the electrical signals d(t)1 to d(t)n.

(41) With this configuration, one canceler unit 30 generates the beat noises from the electrical signals d(t)1 to d(t)n obtained by converting the WDM signal r.sub.s(t), which is an optical signal, and cancels out the beat noises in the electrical signals d(t)1 to d(t)n with the generated beat noises. Thus, it is possible to cancel out the beat noises in the WDM signal r.sub.s(t) and obtain a high-quality signal component. Further, because the canceler unit 30 can be configured using one digital signal processing circuit, the canceler unit 30 can be achieved with low mounting costs.

Modification Example 1 of Embodiment

(42) FIG. 3 is a block diagram illustrating a configuration of a multi-channel compatible interference canceler unit 30A according to Modification Example 1 of the embodiment.

(43) The canceler unit 30A illustrated in FIG. 3 is different from the above-described canceler unit 30 (FIG. 2) in a connection configuration in which first to n-th signals D(t)1a to D(t)na after beat noise cancellation output from the subtraction units 33a to 33n are fed back to input sides of the absolute value calculation units 31a to 31n. This feedback is repeated N times (a plurality of times).

(44) By repeating the feedback N times, the beat noise component can be canceled out so that the beat noise components do not substantially remain in the first to n-th signals D(t)1a to D(t)na.

(45) As described above, when the processing of canceling out the beat noise component of the second term on the right side with the third term in Equation (3b) above is performed only once, a state in which the beat noise component cannot be completely canceled out occurs. In the first cancellation processing, that is, one cancellation processing using the absolute value calculation units 31a to 31n, the scaling units 32a to 32n, and the subtraction units 33a to 33n, the entire right side of Equation (2b) above is substituted into d(t) of |d(t)|.sup.2 in the second term on the right side of Equation (3a). Thus, the beat noise components cannot be completely canceled out in some cases.

(46) Thus, when the first to n-th signals D(t)1a to D(t)na are fed back to the input sides of the absolute value calculation units 31a to 31n, the entire right side of Equation (3b) is substituted into d(t) of |d(t)|.sup.2 in the second term on the right side of Equation (3a) in second cancellation processing.

(47) The second cancellation processing is different from the first cancellation processing in that the equation substituted into d(t) of |d(t)|.sup.2 in the second term on the right side of Equation (3a) does not include a third term on the right side of Equation (3b) in the first cancellation processing, but includes the third term on the right side of Equation (3b) in the second cancellation processing. That is, in the second and subsequent feedbacks, there is the third term on the right side of Equation (3b). This third term serves to gradually cancel out the beat noise component. That is, as the feedback is repeated two times, three times, and so on, the beat noise component is gradually canceled out.

(48) Thus, the beat noise components that remain slightly in the first to n-th signals D(t)1a to D(t)na output from the subtraction units 33a to 33n are fed back, and the feedback beat noise components are further subtracted from the beat noise components in the input electrical signals d(t)1 to d(t)n and canceled out. Through the second cancellation processing and subsequent cancellation processing, it is possible to further reduce the beat noise components remaining in the first cancellation processing.

Modification Example 2 of Embodiment

(49) FIG. 4 is a block diagram illustrating a configuration of a multi-channel compatible interference canceler unit 30B according to Modification Example 2 of the embodiment.

(50) The canceler unit 30B illustrated in FIG. 4 includes demodulation units 41a to 41n, determination units 42a to 42n, transmission source restoration units 43a to 43n, absolute value calculation units (second absolute value calculation units) 44a to 44n, synchronizing and scaling units 45a to 45n, and subtraction units (second subtraction units) 46a to 46n, as illustrated in a broken line frame 30b, in addition to the components of the canceler unit 30 (FIG. 2) described above. Here, the canceler unit 30B has a connection configuration in which the electrical signals d(t)1 to d(t)n are directly input to the subtraction units 46a to 46n at a last stage.

(51) The canceler unit 30B includes n systems with the same circuits that process the n electrical signals d(t)1 to d(t)n. Thus, characteristics of the embodiment will be described with, as a representative, a circuit of a first system that processes the first electrical signal d(t)1. That is, the characteristics will be described with, as a representative, a circuit including the absolute value calculation unit 31a, the scaling unit 32a, the subtraction unit 33a, the demodulation unit 41a, the determination unit 42a, the transmission source restoration unit 43a, the absolute value calculation unit 44a, the synchronizing and scaling unit 45a, and the subtraction unit 46a, which is the circuit of the first system.

(52) The demodulation unit 41a demodulates the signal D(t)1, which is a signal component after beat noise cancellation output from the subtraction unit 33a. However, the beat noise component may remain in the signal D(t)1 output from the subtraction unit 33a as described above.

(53) The determination unit 42a determines a voltage level of a transmission source signal by comparing a voltage level of the demodulated signal with a predetermined threshold value, and determines a signal component with no beat noise of the transmission source (referred to as a transmission source signal component).

(54) The transmission source signal is transmitted in some discrete voltage patterns such as 5 V, 10 V, and 15 V. A beet noise is included in this transmission source signal and, for example, the transmission source signal transmitted at 5 V has a voltage of 4.5 V, 5.5 V, or the like due to superposition of the beat noise or has a voltage of 8 V or the like due to the superposition of beat noise when fluctuation is large. Such a fluctuating signal is transmitted.

(55) The determination unit 42a sets, for example, 7.5 V, which is a middle between 5 V and 10 V, as the threshold value when the determination unit 42a performs the determination using the threshold value. The determination unit 42a determines that the transmission source signal has been transmitted at 10 V when the voltage level exceeds the threshold value of 7.5 V, and determines that the transmission source signal has been transmitted at 5 V when the voltage level is 7.5 V or less. This determination is referred to as a voltage threshold value determination processing.

(56) In practice, there is a case in which the determination is erroneous because the beat noise is included in the transmission source signal as described above. For example, when the transmission source signal has been transmitted at 5 V, but has had a voltage of 8 V due to the beat noise included in the transmission source signal, the determination unit 42a may erroneously determine that the transmission source signal is a transmission source signal that has been transmitted at 10 V. However, the erroneous determination is extremely rare and is within an allowable error range for appropriately determining the transmission source signal component.

(57) A signal for which a transmission voltage level has been determined through the voltage threshold value determination processing by the determination unit 42a is only a transmission source signal component with no beat noise component.

(58) The transmission source restoration unit 43a performs processing of generating an appropriate transmission source signal with no beat noise of the transmission source according to the transmission source signal component.

(59) The absolute value calculation unit 44a performs the absolute value squaring calculation processing on the appropriate transmission source signal to generate an accurate beat noise component, and generates a signal component including this accurate beat noise component. That is, the accurate beat noise component is a beat noise component that is relevant only to the appropriate transmission source signal.

(60) The synchronizing and scaling unit 45a synchronizes the signal component including the accurate beat noise component with the electrical signal d(t)1 and multiplies the accurate beat noise component by the proportional constant through scaling processing to restore the beat noise relevant to the appropriate transmission source signal. This beat noise is input to the respective subtraction units 46a to 46n. That is, n types of beat noises are input to the one subtraction unit 46a.

(61) The subtraction unit 33a subtracts the n types of beat noises from the first electrical signal d(t)1 to cancel out all the beat noises and obtain the first signal D(t)1b of only the signal component.

(62) With the canceler unit 30B having such a configuration, the transmission source signal component with no beat noise of the transmission source is temporarily determined through the voltage threshold value determination processing, and the accurate beat noise component is generated from the determined transmission source signal component. Because the beat noise in the electrical signal d(t)1 is canceled out on the basis of the accurate beat noise component, it is possible to obtain the signal D(t)1b with no beat noise.

(63) In addition, a specific configuration can be changed appropriately without departing from the gist of the present invention.

REFERENCE SIGNS LIST

(64) 10C Optical reception device 11a to 11n Optical hybrid unit 12a to 12n Local light source 13I, 13Q PD 15a to 15n Demodulation unit 30, 30A, 30B Multi-channel compatible interference canceler unit (canceler unit) 31a to 31n Absolute value calculation unit 32a to 32n Scaling unit 33a to 33n Subtraction unit 41a to 41n Demodulation unit 42a to 42n Determination unit 43a to 43n Transmission source restoration unit 44a to 44n Absolute value calculation unit (second absolute value calculation unit) 45a to 45n Synchronizing and scaling unit 46a to 46n Subtraction unit (second subtraction unit)